Al-steel Using Laser

ARTICLE IN PRESS

0030-3992/$ - se

doi:10.1016/j.op

�CorrespondE-mail addr

Optics & Laser Technology 39 (2007) 652–661

www.elsevier.com/locate/optlastec

Dissimilar material joining using laser(aluminum to steel using zinc-based filler wire)

Alexandre Mathieua,�, Rajashekar Shabadib, Alexis Deschampsb, Michel Sueryc,Simone Matteıa, Dominique Greveya, Eugen Cicalaa

aLTm, laboratoire Laser et Traitement des materiaux, IUT Le Creusot, 12 rue de la fonderie, 71200 Le Creusot, FrancebLTPCM, Laboratoire de Thermodynamique et de Physico-Chimie Metallurgiques, ENSEEG, 38402 St. Martin d’Heres, France

cGPM2, laboratoire Genie Physique et Mecanique des Materiaux, ENSPG, 38402 St. Martin d’Heres, France

Received 13 April 2005; received in revised form 29 August 2005; accepted 31 August 2005

Available online 24 October 2005

Abstract

Joining steel with aluminum involving the fusion of one or both materials is possible by laser beam welding technique. This paper

describes a method, called laser braze welding, which is a suitable process to realize this structure. The main problem with thermal joining

of steel/aluminum assembly with processes such as TIG or MIG is the formation of fragile intermetallic phases, which are detrimental to

the mechanical performances of such joints. Braze welding permits a localized fusion of the materials resulting in a limitation on the

growth of fragile phases. This article presents the results of a statistical approach for an overlap assembly configuration using a filler wire

composed of 85% Zn and 15% Al. Tensile tests carried on these assemblies demonstrate a good performance of the joints. The fracture

mechanisms of the joints are analyzed by a detailed characterization of the seams.

r 2005 Elsevier Ltd. All rights reserved.

Keywords: Laser beam joining; Aluminum; Steel

1. Introduction

New European anti-pollution and energy saving laws,which will become effective in 2008 will impose theautomotive industry some requirements concerning reduc-tion in fuel consumption. These requirements can befulfilled partly by reducing the total weight of the vehicles.Because of the easy recyclability of aluminum, there areefforts towards making an all aluminum vehicles. Eventhough the achievable weight reduction reaches 50% theindustrialization of such a vehicle remains a very costlysolution, whereas the introduction of aluminum compo-nents in a standard conception of steel car body is anattractive compromise between cost and performance.Nevertheless, this solution requires assembling steel withaluminum. Among the direct methods of steel/aluminumassembly, one can mention arc-welding [1], plating

e front matter r 2005 Elsevier Ltd. All rights reserved.

tlastec.2005.08.014

ing author. Tel.: +333 85 73 10 56; fax: +33 3 85 73 11 20.

ess: [email protected] (A. Mathieu).

followed by brazing [2], and, ultrasonic welding [3].However, these processes are not suitable for automotiveproduction and for steel/aluminum continuous assemblybecause they are difficult to implement in this context.Aluminum and steel, due to their poor miscibility show a

poor metallurgical compatibility, which promotes theformation of brittle intermetallic phases. However, thedevelopment of new welding techniques in the solid ormushy phase (friction stir welding, welding by explosion)makes possible this sort of assembly providing a compleximplementation. Mechanical assembling techniques such asriveting allow an assembly of different materials but theyare very expensive. Thus, most currently car parts made ofaluminum are those, which do not require direct assemblywith steel, such as openings (cap or tailgate).However, the existing literature shows that steel/alumi-

num assemblies can be obtained by laser beam welding. Inparticular, one can mention the works of the BremerInstitute fur Angegwandte Strahltechnik (BIAS) concern-ing the assembly of dissimilar metals such as steel and

ARTICLE IN PRESS

Fig. 1. Overlap joint configuration.

A. Mathieu et al. / Optics & Laser Technology 39 (2007) 652–661 653

aluminum [4,6–8]. They used a technique called transpar-ency laser braze welding to assemble the steel upon thealuminum. The laser beam heats the steel to a temperaturejust below the melting point, which in turn melts thealuminum sheet by thermal conduction, which is placedjust below the steel sheet. Other works, which are near tothe ideology of the BIAS, combine laser heating and rollwelding to join low-carbon steel and AA-5052 [9]. We callbraze welding an operation that consists of assembling twometals in a permanent manner with or without a fillermetal. A well-known example of laser braze welding is thesteel/steel assembly with a Cu–Si filler wire [10,11]. Anothermethod consists of welding steel on aluminum by usinglaser keyhole welding. This method has been developed byKatayama and Mizutani [12] and more recently by Sierra etal. [13], and found that an efficient joint could be producedto realize the high strength joints between aluminum andsteel using laser keyhole welding. Laser braze welding ofsteel with aluminum has been realized also in theUniversity of Bayreuth [5]. It has been shown that themechanical performances of these joints are within 20%the performances of the classical brazed aluminum joints.

The process presented in this article has many advan-tages. The use of the laser allows a local energy inputresulting in a limitation of the formation of brittle phases.Moreover, the speed of laser processes is generally higherthan that of other thermal processes. The use of the lasertechnique also offers the possibility of new design, which isimpossible with conventional welding processes (such asspot welding, riveting, etc.). In particular, the welding ofhollow section profiled sheets becomes possible thanks tothis mono-access laser process.

However, steel aluminum laser braze welding presentssome difficulties. In certain cases, the obtained assemblieshave a low mechanical performance due to the formationof a fragile intermetallic layer at the steel/seam interface.The growth of these phases depends on the composition ofthe filler wire and the time/temperature history of theprocess, which varies with location in the weld. Manypublications deal with these intermetallic phases formedduring steel/aluminum assembly [4–8]. Kreimeyer andSepold [8] have shown that for a layer thickness below10 mm, the assemblies present high interface strength. In thefirst part of this article, the laser process is described. Anoptimization method of the operating parameters isdescribed. Finally, characterizations of the realized assem-blies are presented in order to understand the causes offracture.

2. Laser braze welding process with filler wire

2.1. Braze welding configuration

This article presents the steel/aluminum assembly for anoverlap joint configuration (Fig. 1). The filler material wasin the form of wire with a diameter of 1.6mm. The wirefeeding speed is close to the brazing speed. The filler

material is chosen taking into account of the metallurgicalcompatibility with aluminum and a melting temperaturecompatible with that of aluminum (Tfusion ¼ 650 1C). Thewire melting is directly caused by the laser beam irradia-tion. The beam size lies between 1 and 2mm. The filler wireis pulled under the laser beam so that the laser beam neverdirectly lights the base parts of the assembly. The processneeds a precise positioning of the laser beam in the jointplane; the filler wire is guided by the joint geometry. Agaseous shielding is brought behind the molten pool inorder to limit the oxidation. The protection gas is a mixtureof 70% Helium and 30% Argon. A 10mm diameter tubewith a flow rate of 20 L/min brings this gas flow.

2.2. Materials

The material for the aluminum part of the assembly ismainly a 6016 T4 aluminum alloy. Partial recrystallizationhas been observed through out the material geometry.Inclusions consisting of Al–Fe and Si have been widelyobserved, as it is generally the case in aluminum products.The grain sizes are about 160–180 mm. The steel sheet isGXES low carbon steel coated with zinc and was of0.77mm thickness. The composition of each material isgiven in Table 1.The zinc layer is obtained by hot dip galvanizing. The

microstructure of this steel is mainly ferritic with the grainsize varying from 60 to 80 mm. The filler material is a zincbase alloy with 15% Al. The composition (in wt%) of thisalloy is given in Table 2. Its density is 5.73 g/cm3 and itsmelting temperature is about 440 1C. This wire is used forlow temperature brazing of aluminum with itself, copperand steel. It consists of two closely bonded phasescombining the property of zinc cathodic protection onthe steel and the properties of chemical and mechanical

ARTICLE IN PRESS

Table 2

Variables and levels (4 with 4 levels and 3 with 2 levels)

P (kW) Def. (mm) a (1) SBW (m/min) SFW (m/min) + (mm) Optic

1.4 +10 35 2.0 2.0 400 Two-spots

1.6 +11 45 2.4 2.4 600 One-spot

1.8 +12 2.8 2.8

2.0 +13 3.2 3.2

Fig. 2. Spatial energy distribution obtained for the two optic shaping

devices (arbitrary units, a and b defocused +10mm).

Table 1

Chemical composition of the materials (wt%)

Material Si Mg Fe Zn Mn Cu Ti Cd Pb Cr Al S P C

AA-6016 1.0–1.5 0.25–0.6 o0.5 o0.2 o0.2 o0.2 o0.15 — — o0.1 Bal. — — —

GXES steel 0.006 — Bal. — 0.105 — 0.068 — — — 0.034 0.006 0.009 0.002

Zn-based filler — — o0.02 Bal. — o0.003 — o0.003 o0.003 — 14–16 — — —

A. Mathieu et al. / Optics & Laser Technology 39 (2007) 652–661654

stability of aluminum. Moreover, its melting temperature iscompatible with the considered application.

2.3. Laser conditions

The laser used for this work is a continuous waveNd:YAG with a maximal power of 3.5 kW. A four-axisdisplacement machine with numerical control carries outthe spatial displacement of the assembly. The laser beam isinjected into an optical fiber diameter 600 mm diameterwhich is brought on the target. Between the end of theoptical fiber and the target, the beam is shaped owing to aclassical system with a magnification of 1. Then the beam isdefocused in order to obtain a spot size between 1 and2mm. The spatial energy distribution is slightly degradedcompared to that obtained when the focal plane is at thesample surface. A laser beam shaping with two spots is alsoused. For this, a prism is inserted between the collimatingand the focusing lenses. The resulting beam consists of twohalf-spots separated to one from the other. Fig. 2 presentsthe energy spatial distribution for both the beam-shapingdevices. The beam diameter is measured at 86% ofmaximum energy, which corresponds to the horizontalline in Figs. 2a and b. In the case of Fig. 2a, the measureddiameter is around 1.6mm and thus close to the used fillerwire diameter.

2.4. Tensile tests

For each trial of welding, four tensile samples have beentested. The tensile sample length is 30mm in width and212mm in length. The loading direction is perpendicular tothe weld line. Since the thicknesses of the sheets are not thesame, the ultimate strength is obtained by dividing theforce at fracture of the specimen by the length of the weldline (30mm) and termed as resistance (N/mm).

3. Optimization parameters

3.1. Design of experiment

Previous experiments have shown that the mechanicalperformances of the joints depend on the operatingparameters. So, in order to optimize the process, a seriesof samples has been made according to the ‘‘TaguchiDesign method’’ [16–18]. The chosen variable were thelaser power (P), defocusing length (def.), tilt angle of theassembly with respect to the laser beam axis (a) (see Fig. 1),braze welding speed (SBW), filler wire speed (SFW),diameter of the fiber (Ø) and the laser beam shaping(one-spot or two-spots). Table 3 gives the parameters andtheir associated levels. The levels of the variables have beendetermined from previous experiences of the other works.From these variables and their associated levels, a design

of experiment composed of 16 tests [18] can be constructed.The answer function considered for each test is themechanical strength of the joint (Flin). The effects of thedifferent variables on the mechanical strength are shown inFig. 3. From Fig. 3, it is learnt that in the investigatedrange the fiber diameter has a negligible influence on the

ARTICLE IN PRESSA. Mathieu et al. / Optics & Laser Technology 39 (2007) 652–661 655

mechanical strength. The other factors can be listed asfollows in the order of decreasing influence: laser power,filler wire speed, braze welding speed, tilt angle, defocusinglength and laser beam shaping. Therefore, Table 3 lists theoptimal conditions found using this method.

3.2. Correlation between mechanical strength and fracture

places

Based on the various experiments carried out for theoptimization of the welding conditions, several fractureplaces were observed. They are given in Table 4 togetherwith the typical values of the fracture strength. It is clearthat there is a close correlation between the fracturestrength and where the fracture occurs: fracture in steelleads to the highest strength, where as fracture across theseam corresponds to the lowest values. Correspondingmacrographs of seams after the fracture are presented inFig. 4. The operating parameters are: laser power of1500W, scanning speed of 2M/min and filler wire feeding

Table 3

Optimized welding parameters

Laser power 1.6–2 kW

Filler wire speed 2–2.4m/min

Braze welding speed 2–2.4m/min

Inclination angle 351

Defocusing (+10; +11)mm

The laser beam optic shaping Two-spots

Fig. 3. Effects of different variables on the mech

Table 4

Relation Flin–fracture localization

Fracture localization Seam Al/Seam S

Flin (N/mm) p150 140–190 1

F

(A

u

th

speed of 2M/min. Both the laser beam shapings have beentested. Fracture occurred at the steel/seam interface forseam a and in the heat-affected zone (HAZ) of the base

anical resistance (N/mm) of the joints.

t/Seam Mixed 6016 HAZ Steel

60–230 190–230 200–245 X245

ig. 4. Macrographs showing the cross-sections of seam after the fracture:

) weld made using defocused +10mm single beam, and (B) weld made

sing defocused +10mm two-spots beam, with the two spots one behind

e other compared to the direction of displacement.

ARTICLE IN PRESS

Fig. 5. Fracture observed in the steel for a joint realized with optimal

parameters (sample width 30mm).

Fig. 6. Back-scattered electrons image of a cross-section at the seam.

Fig. 7. The three zones of the steel/seam interface.

Fig. 8. Steel/seam interface near the seam head as observed at higher

magnification.

A. Mathieu et al. / Optics & Laser Technology 39 (2007) 652–661656

aluminum for seam b (Fig. 4). It was found that the beamshaping has the substantial impact on the occurrence of thefacture. The two-spot shaping seems to be quite good forjoints of this configuration. As shown in Fig. 4, when usingtwo spots beam shaping, weld are always more resistantand fracture occurs in the HAZ. The impact of the two-spots beam shaping is a different energy repartition fromsingle spot. The maximum energy density is lower and thetotal surface irradiation is bigger. All that leads to a softerand longer heating of the weld. For optimized processparameters, fracture is observed in the steel as shown inFig. 5.

4. Microstructure observations

In order to understand the mechanisms of failures,scanning electron microscopy observations and micro-hardness profiles have been carried out.

4.1. Seam

Fig. 6 shows a back-scattered electron micrograph of ajoint. The observed contrasts are due to the differences ofdistribution of the main elements, i.e., Zn, Fe and Al. Thelightest zones correspond to the Zn-richest zones whereasthe darkest ones correspond to the Al-rich zones. There-fore, the composition of the seam is heterogeneous with amean composition of 63% of Al and 37% of Zn (at%). Thedarkest regions near the base aluminum are richer inaluminum. The lightest zones far from the base aluminumare richer in Zinc.

4.2. Steel/seam interface

Three different zones are identified along the steel/seaminterface: the head, the intermediate part and the foot(Fig. 7). The head of the seam (Fig. 8) presents an interface

free of reaction layer indicating that the liquid has simplywetted the steel sheet. It has to be pointed out that thegalvanized layer of the steel has been dissolved in thisregion. Some dendrites seem to initiate from the interfaceand micro-shrinkage pores are observed in the inter-dendritic regions.The intermediate part presents an interface with a

reaction layer with a maximal thickness of 10 mm (Fig. 9).This reaction layer is composed of columnar crystals of

ARTICLE IN PRESSA. Mathieu et al. / Optics & Laser Technology 39 (2007) 652–661 657

variable composition: Al from 63% to 73%, Fe from 18%to 28% and Zn from 5% to 15% (at%). Their growth canbe explained by a mechanism of Fe dissolution andprecipitation into crystals of Fe–Al and Fe–Al–Zncomposition. The compounds in this reaction layer aremainly intermetallic phases like FeAl3 and Fe2Al5 [8]. Thedetected zinc is mainly in solid solution with Al. Fig. 10shows a micro-hardness profile across the steel/seaminterface in the intermediate part as defined earlier inFig. 7. It can be seen that the micro-hardness values of thereaction layer formed between the steel sheet and the seamis around 400Hv. This reaction layer is much harder thanthe steel sheet and the seam (between 120 and 150Hv).

Fig. 9. Steel/seam interface near the intermediate part.

Fig. 10. Hardness profile across the steel/seam interface.

Fig. 11. Steel/seam interface near the seam foot.

The interface zone situated near the seam foot is themost complex. One distinguishes a reaction layer ofthickness below 5 mm. This layer is located between thesteel and a light strip which is made almost of pure Zinc(Fig. 11). One also notes the presence of micro-shrinkagepores close to the interface between the reaction layer andthe light strip. The liquid part of the seam that solidifies last(Zn-rich zones) contracts and thus leads to the formationof the defect. These defects can have more or lessimportance according to the cooling rate. In the caseshown in Fig. 11, these defects are very small and thereforeshould not influence the monotonic mechanical strength.Their effect could however be different for fatigue tests.The presence of the Zn-rich light strip can be explained bya diffusion mechanism of Al in the solid or mushy statetowards the reaction layer from the Zn–Al rich seam. Thisdiffusion explains the Al depletion and therefore the Znenrichment of the zone located near the reaction layer. Themechanism involving the creation of the intermetalliccompound layer (precipitation dissolution) is not the sameas in the case of the intermediate part, since theintermetallic layer in the seam foot does not have thesimilar microstructure. The microstructure near the inter-mediate region has lesser defects and appears more intact.It is possible that a diffusion mechanism of the Al throughthe solid Fe competes with the previous mechanismbecause of the different cooling rate in the seam foot.Fig. 12 presents the fracture surfaces of one sample,

shown on the seam side that breaks fractured at the steel/seam interface. From Fig. 12, it can be observed that thefracture surface composed of cracked and scaled slabs andof small dendritic regions. These slabs come from thereaction layer formed during the process. The brittleness ofthe reaction layer is probably the origin of the fracture inthe joint interface with steel but the low cohesion betweenthe dendritic regions and the steel sheet can explain the lowvalues of the joint bonding strength.

4.3. Seam/aluminum interface

Infiltration of Zn in HAZ by grain boundary liquation isobserved as shown in Fig. 13. This phenomenon iscommonly observed for aluminum alloys [14] and it canexplain why some samples break in the aluminum HAZ ornear the Aluminum/seam interface. Fig. 14 shows a micro-hardness profile across the aluminum/seam interface andsituated approximately at the middle of the aluminumsheet thickness. A significant hardening can be detected inthe brazing zone, which is due to the presence of the hardereutectic mixture. Just beside the seam, a softer region isdetected in aluminum base. The fracture surface of asample which has failed in the aluminum HAZ is shown inFig. 15. It is typical of a ductile fracture that occurs in thisalloy. The fracture occurring at the seam/aluminuminterface could be initiated by the Al–Fe–Si inclusions thatare generally present in the 6xxx aluminum alloys. Most ofthese intermetallic phases of Al–Fe–Si composition are

ARTICLE IN PRESS

Fig. 13. Back scattered electron micrographs of the seam/aluminum interface.

Fig. 12. Fractograph of the seam side of the fracture, when fracture occurred at the HAZ of the aluminum base material.

Fig. 14. Hardness profile across the seam/aluminum interface.

A. Mathieu et al. / Optics & Laser Technology 39 (2007) 652–661658

short rods perpendicularly oriented to the rolling direction(Fig. 15).

4.4. Conclusions of microscopic observations

From the microscopic observations it appears that theformation of intermetallic phases characterized by theirhigh hardness values at the steel/seam interface are not thelonely parameter controlling the strength and ductility ofthe Al/St joints. Indeed, some regions with bad cohesionwith the steel sheet and shrinkage pores are observed at thesteel–seam interface and are probably detrimental to thejoint bonding strength. The presence of this reaction layeris actually necessary to ensure a good cohesion between the

ARTICLE IN PRESS

Fig. 15. Fracture surface observed by SEM, side aluminum fractographs of the aluminum side of the fracture, when fracture occurred at the HAZ of the

aluminum base material.

Fig. 16. Relation between the length of the steel/seam interface and the mechanical resistance of the tensile specimen.

A. Mathieu et al. / Optics & Laser Technology 39 (2007) 652–661 659

seam and the steel sheet. The literature [8] said that below athickness of 10 mm, the reaction layer seems to be not toodetrimental to the mechanical properties. In the presentobservations, the reaction layer was measured to be lessthan 15 mm. Moreover, this reaction layer thickness is notconstant along the steel/seam joint since it can vary from 0to 15 mm.

Fig. 17. Seam geometry and corresponding values of the mechanical

strengths achieved.

5. Correlation between macroscopic observations and the

mechanical strength

5.1. Length of the steel/seam interface

Fig. 16 shows the relation between length of the steel/seam interface and mechanical strength of the tensilespecimens. It clearly appears that the length of the steel/seam has a positive influence on the mechanical strength ofthe sample, which increases linearly as the length of thesteel/seam interface increases. The large scatter of theresults is explained by the fact that for each experimentalpoint reported in Fig. 16, the welding conditions aredifferent.

5.2. Seam geometry

Fig. 17 shows different seam geometries correspondingto different values of mechanical strength achieved. In lightof this, it appears quite clear that the mechanical strength iscorrelated to the seam geometry. The wetting angle of theseam seems to have a strong influence on the mechanicalstrength of the assembly. In order to quantify the

ARTICLE IN PRESS

Fig. 18. Definition of the parameters of the geometrical criterion.

Fig. 19. Relation between the geometrical criterion and the mechanical

strength (Flin).

A. Mathieu et al. / Optics & Laser Technology 39 (2007) 652–661660

correlation between the wetting angle, brazed length andmechanical strength a geometrical criterion based on theratio between the brazed length (L) and the wetting angle(y) as defined in Fig. 18 is proposed.

Fig. 19 presents the variation of the mechanical strengthas a function of the ðL=yÞ ratio. It can be observed thatincreasing the ratio leads to an increase of the mechanicalstrength of the joint. This criterion can be used to predictthe strength of the joint without testing it but simply bymeasuring the values of the two parameters L and y.

6. Conclusion

Heterogeneous steel–aluminum assemblies have beenrealized by laser braze welding. The obtained joints are theresult of welding in the base aluminum and brazing on thesteel sheet. The filler material is a Zn-base alloy chosen forits compatibility with the aluminum alloys and its lowmelting temperature.

A very important aspect of the process is that allexperiments are implemented without the use of a brazing

chemical flux. The laser braze welding process presentsseveral advantages compared to conventional assemblingprocesses such as riveting or spot-welding. Notably, thework speed is raised and the laser process can beautomated and controlled [19].By means of a ‘‘Design of Experiment’’ inspired by

‘‘Taguchi Methods’’, it is possible to find the best operatingparameters window. Under the optimal conditions, themechanical performances of the steel/aluminum assembliesreach fracture strengths superior to 200N/mm. In certaincondition, the rupture occurs in the heat-affected zone(HAZ) of the aluminum or in the steel sheet itself. Themechanical strength values are compatible with thespecifications relative to these assemblies in the automotiveindustry.The causes of failures of the realized joints have been

investigated. It appears that the formation of intermetallicphases characterized by their high hardness values at thesteel/seam interface is not the only cause of the Al/St jointbrittleness, especially when their thickness is below 10 mm.It also appears that the global geometry of the joints(concavity, wetting, etc.), is a significant factor to take intoaccount. This geometry, as well as the formation of theintermetallic compound layer, is governed by the tempera-ture history during the process. Therefore, thermalmodeling and temperature control are necessary steps foran optimal control of the process. Investigations on thetemperature control using an infrared imaging systemare in progress. Until now, the thermal imaging allowedus to validate the results obtained by FEM thermalsimulation [15].

Acknowledgements

This work has been carried out in the project ‘‘A3FL’’ ofthe French Materials and Processes National Network(RNMP: ‘‘Reseau National Materiaux et Procedes’’). Theauthors would like to thank the French ministry ofresearch for the financial support, industrial partners(Renault, Alcan and ARES) for providing the materialsand facilities, and especially Dr. V. Sorel and Dr. I.Bordesoules for fruitful discussions. Academic partners ofthe project (GEMPPM, LMI and LTDS) are also thankedfor fruitful discussions.

References

[1] Murakami T, Nakata K, Tong H, Ushio M. Assemblage mixte acier-

alliage d’aluminum soudo-brase en MIG sur assemblage a clin. Rev

Trans JWRI 2003;32(1).

[2] Roulin M, Luster JW, Karadeniz G, Mortensen A. Strength and

structure of furnace-brazed joints between aluminum and stainless

steel. Weld J: Res Suppl 1999:151-s–5-s.

[3] Tsujino J, Hidai K, Hasegawa A, Kanai R, Matsuura H, Matsushima

K, et al. Ultrasonic butt welding of aluminum, aluminum alloy and

stainless steel plate specimen. Ultrasonics 2002;40:371–4.

[4] Schubert E, Zerner I, Sepold G. Laser beam joining of material

combinations for automotive applications. Proc SPIE 1997;3097:

212–20.

ARTICLE IN PRESSA. Mathieu et al. / Optics & Laser Technology 39 (2007) 652–661 661

[5] Bergmann HW, Waldmann H, Guyenot M, Haldenwanger HG,

Korte M. Untersuchungsergebnisse LaserschweiXen zum Al von Stahl

auf AuXenhautanwendungen, European Automotive Laser Applica-

tion, Bad Nauheim, Germany, Automotive Circle International, 2000.

[6] Schubert E, Klassen M, Zerner I, Walz C, Sepold G. Light-weight

structures produced by laser beam joining for future applications in

automotive and aerospace industry. Newspaper Mater Process

Technol 2001;115:2–8.

[7] Wagner F, Zerner I, Kreimeyer M, Seefeld T, Sepold G, Character-

ization and properties of dissimilar metal combinations of Fe/Al and

Ti/Al sheet materials. In: Proceedings of the ICALEO’01, Jackson-

ville, USA, 2001 (CD-ROM).

[8] Kreimeyer M, Sepold G. Laser steel joined aluminum-hybrid

structures. In: Proceedings of the ICALEO’02 (CD-ROM), Jackson-

ville, USA, 2002.

[9] Rathod MJ, Kutsuna M. Joining of aluminum alloy 5052 and low-

carbon steel by laser roll welding. Weld J: Res Suppl 2004:16-s–26-s.

[10] Hoffman P, Kugler P, Schwab J. Laser brazing with high power solid

state lasers—systems and applications in automotive industry. In:

Proceedings of the second international WLT-conference on lasers in

manufacturing, Munich, Germany, 2003.

[11] Larson JK. Overview of joining technologies in the automotive

industry. Sweden: Volvo Car Corporation.

[12] Katayama S, Mizutani M. Welding laser of aluminum and steel. In:

Proceedings of the ICALEO’03 (CD-ROM), Jacksonville, USA,

2003.

[13] Sierra G, Peyre P, Deschaux-Beaume F, Stuart D, Coste F, Fras G.

Nd:YAG laser welding of aluminum to low carbon steel. In:

Proceedings of the ICALEO’04 (CD-ROM), San Francisco, USA,

2004.

[14] Huang C, Kou S. Partially melted zones in aluminum welds: solution

segregation and mechanical behavior. Weld J: Res Suppl 2001:9-s–17-s.

[15] Mathieu A, Matteı S, Rodriguez L, Deschamps A, Suery M, Viala

JC, et al. Laser brazing of steel/aluminum assembly. In: Proceedings

of the ICALEO’04 (CD-ROM), San Francisco, USA, 2004.

[16] Montgomery DC. In: Design and analysis of experiments. Singapore:

Wiley; 1991.

[17] Goupy J. In: Dunod, editor. Introduction aux plans d’experiences,

Paris, 2001.

[18] Alexis J. Pratique industrielle de la methode Taguchi—Les plans

d’experiences. AFNOR; 1995.

[19] Wallee O, Contribution a l’etude du soudage d’alliages d’aluminum

par faisceau laser avec fil d’apport. Ph.D. thesis, Universite Louis

Pasteur, Strasbourg, France, 2000.